I have to make three processes A, B and C that use shared memory. A and B write 100 integers in the shared memory, and C reads them and writes them to a binary file. That is what I made, but it doesn't work properly. I include <stdio.h>, <math.h>, <fcntl.h> and <time.h>. How to make it work?
struct sync
{
int n;
int lock;
int generated;
char process;
} *b;
int testandset(int* lockPtr)
{
int oldValue = *lockPtr;
return 0 != oldValue;
}
int main()
{
struct sync buff;
int pid, ppid, fp, i;
srand(time(NULL));
b = (struct sync*)malloc(666);
b->n = 0;
b->lock = 0;
b->generated = 0;
i = 0;
printf("Generating numbers\n");
pid = fork();
if (0 == pid)
{
while (100 >= b->generated)
{
while (testandset(&(b->lock)))
{
}
buff.n = rand() % 1001;
buff.process = 'A';
fp = open("db", O_RDWR | O_APPEND);
if (-1 == fp)
fp = open("db", O_CREAT);
write(fp, &buff, sizeof(struct sync));
close(fp);
b->generated++;
b->lock = 0;
}
}
if (0 < pid)
{
ppid = fork();
if (0 == ppid)
{
while (100 >= b->generated)
{
while (testandset(&(b->lock)))
{
}
buff.n = rand() % 1001;
buff.process = 'B';
printf("No: %d %d \n", ++i, buff.n);
fp = open( "db", O_RDWR | O_APPEND );
if (-1 == fp)
fp = open("db", O_CREAT);
write(fp, &buff, sizeof(struct sync));
close(fp);
b->generated++;
b->lock = 0;
}
}
if (0 < ppid)
{
wait();
i = 0;
fp = open("db", O_RDONLY, 0755);
printf("Reading from file\n");
while (read(fp, &buff, sizeof(struct sync)))
{
if ('A' == buff.process)
i++;
}
close(fp);
int vals[i];
i = 0;
fp = open("db", O_RDONLY, 0666);
while (read(fp, &buff, sizeof(struct sync)))
{
if ('A' == buff.process)
vals[i++] = buff.n;
}
close(fp);
fp = open("db", O_RDONLY, 0455);
int i;
for(i = 0; i < i; i++)
write((const void*) &vals[i],sizeof(int),1,fp);
}
wait();
}
return 0;
}
Files are not really reliable for mutliprocess sharing information (and editing at the same time), You'd rather use a real database in transaction mode or use IPC http://www.cs.cf.ac.uk/Dave/C/node27.html. Or maybe re-design and use threads and mutex.
There are many issues here, but I will concentrate on testandset(). Firstly, it's misnamed, it doesn't set anything, nor do you set b->lock anywhere. What's it for?
It seems to me that you might be trying to use it for some sort of locking between processes, but it won't work. When you fork, the new child effectively gets a complete copy of the parent's address space and changes to variables in the parent will not be seen in the child, and vice versa. As Maresh says, you need to look at using inter process communication.
Related
right now, I am currently trying to output the contents of buf.mtext so I can make sure take the correct input before moving on with my program. Everything seems to work fine, except one thing; msgrcv() puts garbage characters into the buffer, and the reciever process outputs garbage characters.
Here is my sender process:
int main (void)
{
int i; // loop counter
int status_01; // result status
int msqid_01; // message queue ID (#1)
key_t msgkey_01; // message-queue key (#1)
unsigned int rand_num;
float temp_rand;
unsigned char eight_bit_num;
unsigned char counter = 0;
unsigned char even_counter = 0;
unsigned char odd_counter = 0;
srand(time(0));
struct message {
long mtype;
char mtext[BUFFER_SIZE];
} buf_01;
msgkey_01 = MSG_key_01; // defined at top of file
msqid_01 = msgget(msgkey_01, 0666 | IPC_CREAT)
if ((msqid_01 <= -1) { exit(1); }
/* wait for a key stroke at the keyboard ---- */
eight_bit_num = getchar();
buf_01.mtype = 1;
/* send one eight-bit number, one at a time ------------ */
for (i = 0; i < NUM_REPEATS; i++)
{
temp_rand = ((float)rand()/(float)RAND_MAX)*255.0;
rand_num = (int)temp_rand;
eight_bit_num = (unsigned char)rand_num;
if ((eight_bit_num % 2) == 0)
{
printf("Even number: %d\n", eight_bit_num);
even_counter = even_counter + eight_bit_num;
}
else
{
printf("Odd number: %d\n", eight_bit_num);
odd_counter = odd_counter + eight_bit_num;
}
/* update the counters ------------------------------ */
counter = counter + eight_bit_num;
if((eight_bit_num % 2) == 0) { even_counter = even_counter + eight_bit_num; }
else { odd_counter = odd_counter + eight_bit_num; }
buf_01.mtext[0] = eight_bit_num; // copy the 8-bit number
buf_01.mtext[1] = '\0'; // null-terminate it
status_01 = msgsnd(msqid_01, (struct msgbuf *)&buf_01, sizeof(buf_01.mtext), 0);
status_01 = msgctl(msqid_01, IPC_RMID, NULL);
}
Here is my receiver process:
int main() {
struct message {
long mtype;
char mtext[BUFFER_SIZE];
} buf;
int msqid;
key_t msgkey;
msgkey = MSG_key_01;
msqid = msgget(msgkey, 0666); // connect to message queue
if (msqid < 0) {
printf("Failed\n");
exit(1);
}
else {
printf("Connected\n");
}
if (msgrcv(msqid, &buf, BUFFER_SIZE, 0, 0) < 0) { // read message into buf
perror("msgrcv");
exit(1);
}
printf("Data received is: %s \n", buf.mtext);
printf("Done receiving messages.\n");
return 0;
}
The output is usually something like as follows:
Data received is: ▒
Done receiving messages.
I have made sure to clear my message queues each time after running the sender and receiver processes, as well, since I have come to find out this can cause issues. Thanks in advance for your help.
Turns out neither of the suggested solutions were the issue, as I suspected; the sender process actually works just fine. The problem was that I was trying to print buf.mtext instead of buf.mtext[0] which isn't an actual integer value. I fixed the issue by just doing this:
int temp_num = buf.mtext[0];
printf("Data recieved is %d \n", temp_num);
I have quite a lot of debugging monitoring all over my program so whenever something undesired happens a message appears in XCode with "std::cout" showing what happended, where it happened, and so on.
While I was testing the app on an iPhone or iPad connected to my computer, this worked as well (as I always had XCode open to show the fault).
But now I installed the app on devices of several beta-testers and they do not see these messages...
Rewriting the code to route all the "cout" to a string would cost a lot of time as they appear everywhere in several classes and sub classes, etc...
is there a possibility of simply reading out the last line of the output console or detecting the event of writing to the console and then copying it over to a separate string?
This is something I've done on some android projects to forward stdout and stderr to logcat. You could use this same approach to forwards the stdout/stderr to anywhere you want:
struct stream {
const char *name;
int fd[2];
FILE *src;
};
static void*
log_thread(void *arg)
{
struct stream *stream = arg;
char buf[4000], *off = buf, *nl; // Can't be too big or android stops logging
for (ssize_t r = 0;;off += r, r = 0) {
if (off - buf < sizeof(buf) - 1) {
errno = 0;
r = read(stream->fd[0], off, (sizeof(buf) - 1) - (off - buf));
if (r <= 0) { if (errno == EINTR) continue; else break; }
off[r] = 0;
}
if ((nl = strrchr(off, '\n'))) {
*nl = 0; ++nl;
__android_log_write(ANDROID_LOG_INFO, stream->name, buf);
r = (off + r) - nl;
memcpy((off = buf), nl, r);
} else if (off - buf >= sizeof(buf)) {
__android_log_write(ANDROID_LOG_INFO, stream->name, buf);
r = 0; off = buf;
}
}
close(stream->fd[0]);
close(stream->fd[1]);
return NULL;
}
__attribute__((constructor)) static void
log_init(void) {
static struct stream stream[] = { { .name = "stdout" }, { .name = "stderr" } };
stream[0].src = stdout; stream[1].src = stderr;
for (size_t i = 0; i < sizeof(stream) / sizeof(stream[0]); ++i) {
setvbuf(stream[i].src, NULL, _IOLBF, BUFSIZ);
pipe(stream[i].fd);
dup2(stream[i].fd[1], fileno(stream[i].src));
pthread_t thread;
pthread_create(&thread, 0, log_thread, &stream[i]);
pthread_detach(thread);
}
}
I'm trying to solve my school project in C++. I have to create 15 processes and they have to run in order what means that processes run in this order 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0. It work but when I try to remove semaphore from the memory I am getting error from semctl. On the end I use "semctl(semid, 0, IPC_RMID, 0" but I get error 22 which means EINVAL but it doesn't make sense and I try to remove semaphore from parrent process so I should have privileges to do that.
#include <stdio.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/sem.h>
#include <errno.h>
#include <sys/wait.h>
union semun {
int val;
struct semid_ds *buf;
ushort *array;
};
struct sembuf sops[1];
int semid;
int wait_sem(int index, int pid){
fprintf(stderr, "\n------- Proces %d do operation wait (-1) on semaphore %d\n",pid, index);
sops[0].sem_num = index;
sops[0].sem_op = -1;
sops[0].sem_flg = 0 ;
if (semop(semid, sops, 1)<0){
perror("semop fail wait");
return 1;
}
else
return 0;
}
int signal_sem(int index, int pid){
fprintf(stderr, "\n++++++ Proces %d vykonava operaciu signal (1) na semafore %d\n",pid,index);
sops[0].sem_num = index;
sops[0].sem_op = 1;
sops[0].sem_flg = 0;
if (semop(semid, sops, 1)<0){
perror("semop fail signal");
return 1;
}
else
return 0;
}
void createSem(key_t paKey, int paSemFlg, int paNsems)
{
printf ("uid=%d euid=%d\n", (int) getuid (), (int) geteuid ());
(semid = semget(paKey, paNsems, paSemFlg));
for (int i = 0; i < paNsems; ++i) {
semctl(semid, i, SETVAL, 0);
}
}
void kic()
{
printf("\naaaaaaaaaaaaaa\n");
}
int main() {
key_t key = 1234;
int semflg = IPC_CREAT | 0666;
int nsems = 15;
int semid;
fprintf(stderr, "%d=", sops);
createSem(IPC_PRIVATE, semflg, nsems);
if (semid == -1) {
perror("semget: semget failed");
return 1;
}
else
fprintf(stderr, "semget: semget sucess: semid = %d, parrent pid %d\n", semid, getpid());
int PROCESS_ID = 0;
pid_t PID;
for (int i = 1; i < nsems; i++) {
PID = fork();
if(PID == 0)
{
PROCESS_ID = i;
break;
}
}
if(PID == -1)
{
printf("\nPID ERROR");
}
if(PID != 0) //parrent
{
printf("\n\nparrent with ID %d", PROCESS_ID);
signal_sem(PROCESS_ID+1, PROCESS_ID);
wait_sem(PROCESS_ID, PROCESS_ID);
printf ("uid=%d euid=%d\n", (int) getuid (), (int) geteuid ());
printf("\nEND %d\n", getpid());
int s;
wait(&s);
if((semctl(semid, 0, IPC_RMID, 0))==-1)
{
int a = errno;
printf("\nERROR IPC_RMID %d\n", a);
}
}
if(PID == 0)//child
{
if(wait_sem(PROCESS_ID, PROCESS_ID) == 0){
printf("\nI am child with ID %d", PROCESS_ID);
int ID_NEXT_PROCESS = 1+PROCESS_ID;
if(ID_NEXT_PROCESS == nsems)
ID_NEXT_PROCESS = 0;
signal_sem(ID_NEXT_PROCESS, PROCESS_ID);
return 0;
}
}
return 0;
}
You have two semids. One in global scope, another local to main (which shadows global, you should see a warning). createSem only knows about global one, and initializes it. semctl is called directly by main, and is passed the local one, which is garbage.
Am working in a project where i have to read a set of files and put it in a buffer.The List comprises of small as well as large files.I have to read these files and for more efficiency i tried implementing it in multiple threads.Each thread will take a file from vector of file names and start reading it put it into a buffer and these buffer have to be put in a queue.I happened to have some error in program and i i don't know where exactly in my program the error occurs also don't know why ? Please help me whether there is any mistake in my logic or in my code and how to correct it. Thanks in advance
using namespace std;
#define MAX_THREADS 2
#define BUFFER_SIZE 8388608
vector<string>files;
deque<string>bufferq;
CRITICAL_SECTION Readlock;
int count = 0;
DWORD WINAPI ReadThread(LPVOID s);
int main(int argc,char *argv[])
{
HANDLE ReadT[MAX_THREADS];
char *filelist[5];
DWORD threadid;
filelist[0] = "1.txt";
filelist[1] = "cloudy.jpg";
filelist[2] = "connectify.exe";
filelist[3] = "VMware.exe";
filelist[4] = "Sherlock.mp4";
for(int i=0;i<5;i++)
files.push_back(filelist[i]);
InitializeCriticalSection(&Readlock);
long t1 = GetTickCount();
for(int k = 0; k< MAX_THREADS; k++)
ReadT[k] = CreateThread(NULL,0,ReadThread,NULL,NULL,&threadid);
WaitForMultipleObjects(MAX_THREADS,ReadT,TRUE,INFINITE);
cout << " Time Taken "<< GetTickCount()-t1 << "ms" ;
system("pause");
return 0;
}
DWORD WINAPI ReadThread(LPVOID s)
{
long pending = 0;
//int freespace = BUFFER_SIZE;
char *filename = new char[50];
char fsize[10];
string file;
char *buf;
buf = new char[BUFFER_SIZE];
long filesize = 0;
int numfiles = files.size();
int filled = 0;
int i = 0;
FILE *fp;
char* ptr;
ptr = buf;
while(true)
{
EnterCriticalSection(&Readlock);
if(files.empty())
{
LeaveCriticalSection(&Readlock);
break;
}
else
{
file = files.front();
files.erase(files.begin());
LeaveCriticalSection(&Readlock);
}
bool buff_full = false;
buf = ptr;
int freespace = BUFFER_SIZE;
memset(buf,0,BUFFER_SIZE);
if(!buff_full)
{
if(pending == 0)
{
fp = fopen(file.c_str(),"rb");
if(!fp)
{
cout<<"\nNo such file";
cout<<files[i];
system("pause");
return 0;
}
int r1 =fseek(fp, 0L, SEEK_END);
filesize = ftell(fp);
int r2 =fseek(fp, 0L, SEEK_SET);
sprintf(fsize, "%ld", filesize);
if(freespace >= (strlen(fsize) + strlen(file.c_str()) + 2))
{
count++;
memcpy(buf, file.c_str(), strlen(file.c_str())+1);
freespace = freespace - strlen(file.c_str()) - 1;
buf += strlen(file.c_str()) + 1;
memcpy(buf,fsize,strlen(fsize)+1);
buf += strlen(fsize) + 1;
freespace = freespace - strlen(fsize) - 1;
cout<<"Files read is "<<count<<"\n";
if(freespace == 0)
{
buff_full = true;
pending = filesize;
break;
}
}
else
{
filled = BUFFER_SIZE - freespace;
fclose(fp);
break;
}
if(freespace >= filesize)
{
fread(buf, 1, filesize, fp);
buf += filesize;
freespace = freespace - filesize;
bufferq.push_back(buf);
//cout << "pop"<<bufferq.size();
//i++;
if(files.empty())
{
filled = BUFFER_SIZE - freespace;
fclose(fp);
break;
}
fclose(fp);
}
else
{
fread(buf, 1, freespace, fp);
bufferq.push_back(buf);
//cout <<"pop "<<bufferq.size();
buff_full = true;
}
}
else
{
if(freespace >= pending)
{
fread(buf, 1, pending, fp);
bufferq.push_back(buf);
freespace = freespace - pending;
pending = 0;
//i++;
if(files.empty())
{
filled = BUFFER_SIZE - freespace;
fclose(fp);
break;
}
if(freespace > 0)
buf += pending;
else
buff_full = true;
fclose(fp);
}
else
{
fread(buf, 1, freespace, fp);
bufferq.push_back(buf);
cout << bufferq.size();
pending = pending - freespace;
buff_full = true;
}
}
}
if(buff_full)
{
buf = ptr;
cout << "popping buffer " << bufferq.size();
//bufferq.pop_back();
}
}
return 0;
}
In the context that bug occurs on big files, I suppose that this line can cause problems
sprintf(fsize, "%ld", filesize);
fsize is char[10], and if filesize is >= 1,000,000,000 you'll overwrite fsize array with trailing 0. This will cause "Run-Time Check Failure #2 - Stack around the variable 'fsize' was corrupted.", as you wrote. Please check the sizes of your test files.
Among others, you are filling files in loop on i, and then you wrote:
files.erase(files.begin());
// ...
cout<<"\nNo such file";
cout<<files[i];
files[i] already points to another element as you erased them, and if files are empty on the last iteration it will cause crash.
And what for are you copying file and fsize to buf if you do not copy it to the bufferq?
As bufferq is writable and is shared between threads the access to it should be protected by lock, critical section as you chose.
That's my little code review.
I've got the following two programs, one acting as a reader and the other as a writer. The writer seems to only send about 3/4 of the data correctly to be read by the reader. Is there any way to guarantee that all the data is being sent? I think I've got it set up so that it reads and writes reliably, but it still seems to miss 1/4 of the data.
Heres the source of the writer
#define pipe "/tmp/testPipe"
using namespace std;
queue<string> sproutFeed;
ssize_t r_write(int fd, char *buf, size_t size) {
char *bufp;
size_t bytestowrite;
ssize_t byteswritten;
size_t totalbytes;
for (bufp = buf, bytestowrite = size, totalbytes = 0;
bytestowrite > 0;
bufp += byteswritten, bytestowrite -= byteswritten) {
byteswritten = write(fd, bufp, bytestowrite);
if(errno == EPIPE)
{
signal(SIGPIPE,SIG_IGN);
}
if ((byteswritten) == -1 && (errno != EINTR))
return -1;
if (byteswritten == -1)
byteswritten = 0;
totalbytes += byteswritten;
}
return totalbytes;
}
void* sendData(void *thread_arg)
{
int fd, ret_val, count, numread;
string word;
char bufpipe[5];
ret_val = mkfifo(pipe, 0777); //make the sprout pipe
if (( ret_val == -1) && (errno != EEXIST))
{
perror("Error creating named pipe");
exit(1);
}
while(1)
{
if(!sproutFeed.empty())
{
string s;
s.clear();
s = sproutFeed.front();
int sizeOfData = s.length();
snprintf(bufpipe, 5, "%04d\0", sizeOfData);
char stringToSend[strlen(bufpipe) + sizeOfData +1];
bzero(stringToSend, sizeof(stringToSend));
strncpy(stringToSend,bufpipe, strlen(bufpipe));
strncat(stringToSend,s.c_str(),strlen(s.c_str()));
strncat(stringToSend, "\0", strlen("\0"));
int fullSize = strlen(stringToSend);
signal(SIGPIPE,SIG_IGN);
fd = open(pipe,O_WRONLY);
int numWrite = r_write(fd, stringToSend, strlen(stringToSend) );
cout << errno << endl;
if(errno == EPIPE)
{
signal(SIGPIPE,SIG_IGN);
}
if(numWrite != fullSize )
{
signal(SIGPIPE,SIG_IGN);
bzero(bufpipe, strlen(bufpipe));
bzero(stringToSend, strlen(stringToSend));
close(fd);
}
else
{
signal(SIGPIPE,SIG_IGN);
sproutFeed.pop();
close(fd);
bzero(bufpipe, strlen(bufpipe));
bzero(stringToSend, strlen(stringToSend));
}
}
else
{
if(usleep(.0002) == -1)
{
perror("sleeping error\n");
}
}
}
}
int main(int argc, char *argv[])
{
signal(SIGPIPE,SIG_IGN);
int x;
for(x = 0; x < 100; x++)
{
sproutFeed.push("All ships in the sea sink except for that blue one over there, that one never sinks. Most likley because it\'s blue and thats the mightiest colour of ship. Interesting huh?");
}
int rc, i , status;
pthread_t threads[1];
printf("Starting Threads...\n");
pthread_create(&threads[0], NULL, sendData, NULL);
rc = pthread_join(threads[0], (void **) &status);
}
Heres the source of the reader
#define pipe "/tmp/testPipe"
char dataString[50000];
using namespace std;
char *getSproutItem();
void* readItem(void *thread_arg)
{
while(1)
{
x++;
char *s = getSproutItem();
if(s != NULL)
{
cout << "READ IN: " << s << endl;
}
}
}
ssize_t r_read(int fd, char *buf, size_t size) {
ssize_t retval;
while (retval = read(fd, buf, size), retval == -1 && errno == EINTR) ;
return retval;
}
char * getSproutItem()
{
cout << "Getting item" << endl;
char stringSize[4];
bzero(stringSize, sizeof(stringSize));
int fd = open(pipe,O_RDONLY);
cout << "Reading" << endl;
int numread = r_read(fd,stringSize, sizeof(stringSize));
if(errno == EPIPE)
{
signal(SIGPIPE,SIG_IGN);
}
cout << "Read Complete" << endl;
if(numread > 1)
{
stringSize[numread] = '\0';
int length = atoi(stringSize);
char recievedString[length];
bzero(recievedString, sizeof(recievedString));
int numread1 = r_read(fd, recievedString, sizeof(recievedString));
if(errno == EPIPE)
{
signal(SIGPIPE,SIG_IGN);
}
if(numread1 > 1)
{
recievedString[numread1] = '\0';
cout << "DATA RECIEVED: " << recievedString << endl;
bzero(dataString, sizeof(dataString));
strncpy(dataString, recievedString, strlen(recievedString));
strncat(dataString, "\0", strlen("\0"));
close(fd);
return dataString;
}
else
{
return NULL;
}
}
else
{
return NULL;
}
close(fd);
}
int main(int argc, char *argv[])
{
int rc, i , status;
pthread_t threads[1];
printf("Starting Threads...\n");
pthread_create(&threads[0], NULL, readItem, NULL);
rc = pthread_join(threads[0], (void **) &status);
}
You are definitely using signals the wrong way. Threads are completely unnecessary here - at least in the code provided. String calculations are just weird. Get this book and do not touch the keyboard until you finished reading :)
The general method used to send data through named pipes is to tack on a header with the length of the payload. Then you read(fd, header_len); read(rd, data_len); Note the latter read() will need to be done in a loop until data_len is read or eof. Note also if you've multiple writers to a named pipe then the writes are atomic (as long as a reasonable size) I.E. multiple writers will not case partial messages in the kernel buffers.
It's difficult to say what is going on here. Maybe you are getting an error returned from one of your system calls? Are you sure that you are successfully sending all of the data?
You also appear to have some invalid code here:
int length = atoi(stringSize);
char recievedString[length];
This is a syntax error, since you cannot create an array on the stack using a non-constanct expression for the size. Maybe you are using different code in your real version?
Do you need to read the data in a loop? Sometimes a function will return a portion of the available data and require you to call it repeatedly until all of the data is gone.
Some system calls in Unix can also return EAGAIN if the system call is interrupted - you are not handling this case by the looks of things.
You are possibly getting bitten by POSIX thread signal handling semantics in your reader main thread.
The POSIX standard allows for a POSIX thread to receive the signal, not necessarily the thread you expect. Block signals where not wanted.
signal(SIG_PIPE,SIG_IGN) is your friend. Add one to reader main.
POSIX thread handling semantics, putting the POS into POSIX. ( but it does make it easier to implement POSIX threads.)
Examine the pipe in /tmp with ls ? is it not empty ?